chore(release): 0.12.6 [skip ci]

Author: semantic-release

CHANGELOG.md

@@ -2,6 +2,69 @@
 
 <!-- version list -->
 
+## v0.12.6 (2026-09-23)
+
+### Bug Fixes
+
+- **claude-code**: End a turn the CLI did not cap at max_turns
+  ([#194](https://github.com/UiPath/coder_eval/pull/194),
+  [`17fb10f`](https://github.com/UiPath/coder_eval/commit/17fb10f1250b23df0aa80d33e568996b892711dd))
+
+- **claude-code**: Kill the CLI on a cooperative stop
+  ([#194](https://github.com/UiPath/coder_eval/pull/194),
+  [`17fb10f`](https://github.com/UiPath/coder_eval/commit/17fb10f1250b23df0aa80d33e568996b892711dd))
+
+- **claude-code**: Kill the CLI on a max_turns backstop stop without a turn timeout
+  ([#194](https://github.com/UiPath/coder_eval/pull/194),
+  [`17fb10f`](https://github.com/UiPath/coder_eval/commit/17fb10f1250b23df0aa80d33e568996b892711dd))
+
+- **codex**: End a capped turn before the next call can run a tool
+  ([#194](https://github.com/UiPath/coder_eval/pull/194),
+  [`17fb10f`](https://github.com/UiPath/coder_eval/commit/17fb10f1250b23df0aa80d33e568996b892711dd))
+
+- **delegate**: Count a tool-only reply as a model call
+  ([#194](https://github.com/UiPath/coder_eval/pull/194),
+  [`17fb10f`](https://github.com/UiPath/coder_eval/commit/17fb10f1250b23df0aa80d33e568996b892711dd))
+
+- **delegate**: Keep counting model calls when a tool never returns
+  ([#194](https://github.com/UiPath/coder_eval/pull/194),
+  [`17fb10f`](https://github.com/UiPath/coder_eval/commit/17fb10f1250b23df0aa80d33e568996b892711dd))
+
+- **evalboard**: Compare visible turns, not SDK num_turns, against expected_turns
+  ([#194](https://github.com/UiPath/coder_eval/pull/194),
+  [`17fb10f`](https://github.com/UiPath/coder_eval/commit/17fb10f1250b23df0aa80d33e568996b892711dd))
+
+- **pricing**: Add Opus 5.5 and GPT-6, correct rates that moved publicly
+  ([#195](https://github.com/UiPath/coder_eval/pull/195),
+  [`11b902f`](https://github.com/UiPath/coder_eval/commit/11b902fa9bc0d307bcab22801dc56523fcc1d92a))
+
+- **run-limits**: Count max_turns as model API calls on every harness
+  ([#194](https://github.com/UiPath/coder_eval/pull/194),
+  [`17fb10f`](https://github.com/UiPath/coder_eval/commit/17fb10f1250b23df0aa80d33e568996b892711dd))
+
+- **run-limits**: Count max_turns as model API calls on every harness and enforce it on claude-code
+  ([#194](https://github.com/UiPath/coder_eval/pull/194),
+  [`17fb10f`](https://github.com/UiPath/coder_eval/commit/17fb10f1250b23df0aa80d33e568996b892711dd))
+
+- **run-limits**: Stop a turn mid-flight once a token or USD budget is crossed
+  ([#194](https://github.com/UiPath/coder_eval/pull/194),
+  [`17fb10f`](https://github.com/UiPath/coder_eval/commit/17fb10f1250b23df0aa80d33e568996b892711dd))
+
+- **simulation**: End a dialog with agent_max_turns when the agent hits its own cap
+  ([#194](https://github.com/UiPath/coder_eval/pull/194),
+  [`17fb10f`](https://github.com/UiPath/coder_eval/commit/17fb10f1250b23df0aa80d33e568996b892711dd))
+
+### Documentation
+
+- **parity**: Record the claude-code max_turns backstop
+  ([#194](https://github.com/UiPath/coder_eval/pull/194),
+  [`17fb10f`](https://github.com/UiPath/coder_eval/commit/17fb10f1250b23df0aa80d33e568996b892711dd))
+
+- **run-limits**: Say max_turns restarts per iteration and expected_turns counts tool calls
+  ([#194](https://github.com/UiPath/coder_eval/pull/194),
+  [`17fb10f`](https://github.com/UiPath/coder_eval/commit/17fb10f1250b23df0aa80d33e568996b892711dd))
+
+
 ## v0.12.5 (2026-09-22)
 
 ### Bug Fixes

action.yml

@@ -25,7 +25,7 @@ inputs:
   version:
     description: coder-eval version to install from PyPI, or "local" to install from the action checkout
     required: false
-    default: "0.12.5"   # <-- kept in sync with releases by release.yml
+    default: "0.12.6"   # <-- kept in sync with releases by release.yml
   extras:
     description: >-
       Comma-separated coder-eval extras (`codex`, `antigravity,litellm`), composed

plugins/coder-eval/.claude-plugin/plugin.json

@@ -2,7 +2,7 @@
   "$schema": "https://json.schemastore.org/claude-code-plugin-manifest.json",
   "name": "coder-eval",
   "displayName": "Coder Eval",
-  "version": "0.12.5",
+  "version": "0.12.6",
   "description": "Test whether your Claude Code skills actually trigger, and benchmark any coding agent — author, run, and analyze the eval suite that proves it, locally or as a CI gate.",
   "author": { "name": "UiPath", "email": "coder-eval@uipath.com", "url": "https://github.com/UiPath/coder_eval" },
   "homepage": "https://coder-eval.com",

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "coder-eval"
-version = "0.12.5"
+version = "0.12.6"
 description = "Evaluate, benchmark, and A/B-test AI coding agents (Claude Code, Codex, Gemini/Antigravity, OpenCode, Pi, Delegate) with sandboxed, reproducible YAML task suites."
 readme = "README.md"
 license = "Apache-2.0"

src/coder_eval/__init__.py

@@ -1,3 +1,3 @@
 """coder_eval - A framework for evaluating AI coding agents."""
 
-__version__ = "0.12.5"
+__version__ = "0.12.6"